{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 5.490424211005471,
      "learning_rate": 6.25e-07,
      "logits/chosen": -1.73323655128479,
      "logits/rejected": -1.963712453842163,
      "logps/chosen": -64.71795654296875,
      "logps/rejected": -92.56527709960938,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.8759007361583007,
      "learning_rate": 4.989014955054746e-06,
      "logits/chosen": -1.6668554544448853,
      "logits/rejected": -1.8378652334213257,
      "logps/chosen": -75.77650451660156,
      "logps/rejected": -130.23812866210938,
      "loss": 0.6061,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.1238466128706932,
      "rewards/margins": 0.300619512796402,
      "rewards/rejected": -0.4244661331176758,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.726977629562235,
      "learning_rate": 4.614580148147744e-06,
      "logits/chosen": -1.2113852500915527,
      "logits/rejected": -1.529423475265503,
      "logps/chosen": -59.67132568359375,
      "logps/rejected": -267.4389953613281,
      "loss": 0.367,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.013831649906933308,
      "rewards/margins": 1.7786989212036133,
      "rewards/rejected": -1.7648674249649048,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 4.853755500158873,
      "learning_rate": 3.783685483787105e-06,
      "logits/chosen": -1.130287766456604,
      "logits/rejected": -1.5208537578582764,
      "logps/chosen": -86.93096160888672,
      "logps/rejected": -550.50830078125,
      "loss": 0.1014,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.2528189718723297,
      "rewards/margins": 4.348726749420166,
      "rewards/rejected": -4.601546287536621,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.41367369924975356,
      "learning_rate": 2.6756904723632325e-06,
      "logits/chosen": -1.32732355594635,
      "logits/rejected": -1.7840182781219482,
      "logps/chosen": -165.8927001953125,
      "logps/rejected": -817.1434326171875,
      "loss": 0.0122,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0844600200653076,
      "rewards/margins": 6.145598411560059,
      "rewards/rejected": -7.230058193206787,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.989910262766891,
      "learning_rate": 1.5297703683193755e-06,
      "logits/chosen": -1.598076343536377,
      "logits/rejected": -1.9638893604278564,
      "logps/chosen": -143.41690063476562,
      "logps/rejected": -1029.6900634765625,
      "loss": 0.0085,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.8121221661567688,
      "rewards/margins": 8.604238510131836,
      "rewards/rejected": -9.416360855102539,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.05482660827333183,
      "learning_rate": 5.932870547240455e-07,
      "logits/chosen": -1.44638991355896,
      "logits/rejected": -1.9433313608169556,
      "logps/chosen": -140.36776733398438,
      "logps/rejected": -1133.6243896484375,
      "loss": 0.0022,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8202629089355469,
      "rewards/margins": 9.569122314453125,
      "rewards/rejected": -10.389385223388672,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 10.857846321858867,
      "learning_rate": 6.839273668796747e-08,
      "logits/chosen": -1.6353269815444946,
      "logits/rejected": -2.07700514793396,
      "logps/chosen": -172.53036499023438,
      "logps/rejected": -1374.29296875,
      "loss": 0.0072,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0923078060150146,
      "rewards/margins": 11.75459098815918,
      "rewards/rejected": -12.846899032592773,
      "step": 70
    },
    {
      "epoch": 3.0,
      "step": 75,
      "total_flos": 0.0,
      "train_loss": 0.1484553155489266,
      "train_runtime": 7941.2801,
      "train_samples_per_second": 0.604,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 10,
  "max_steps": 75,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}